This invention concerns a capacitive measurement system with several electrodes, that determines characteristic properties of a conductive body, such as shape and position, by means of capacitive coupling between each pair of these electrodes across the electrically conductive body. To this end an ac voltage is applied to each of the electrodes, and the currents flowing from the electrodes are measured. The respective coupling capacitances can then be calculated from the values for the measurement voltages and electrode currents.
Such a measurement system is described in LU-A-88 828 for instance. This measurement system comprises at least one transmit electrode and at least one receive electrode that are capacitively coupled by a conductive body. The receive electrodes are connected to an analysis circuit that determines the capacitive coupling of the transmit antenna with the conductive body by comparing the measured signal with a reference signal.
When measuring the capacitive coupling using an earthed measurement instrument, the capacitance of the conductive body with respect to ground constitutes an error source. Patent application LU 88 828 proposes dc isolation of the circuit to remove this error. Here the signal from the transmit electrode is isolated from the system ground to reduce the effect of the parasitic capacitance between ground and the body being detected. This type of isolation is however relatively expensive and can only be integrated in semiconductor technology with difficulty.
The object of this invention is to present a capacitive measurement method resp. a capacitive measurement system of said type, that works without dc isolation of the individual components from ground.
This objective is met by a method for determining the capacitive coupling of at least two electrodes with an electrically conductive body, wherein an ac voltage is applied to each of the electrodes, and the currents flowing from each of the electrodes are measured. According to the invention the currents flowing from each of the electrodes are measured n times, nxe2x89xa72, wherein different ac voltages are applied to the electrodes for each of the measurements, and the coupling capacitances are calculated in a network analysis using the respective current and voltage values obtained in the n measurements.
Making multiple measurements with different measurement voltages enables an equation system to be set up using network analysis that can be resolved into the different capacitances (coupling capacitances and parasitic capacitances). Determining the coupling capacitances of two electrodes by taking two measurements with different measurement voltages results from Kirchhoff""s laws in an equation system with 6 equations, for example, from which the coupling capacitances of interest and the parasitic capacitances as well as the parasitic currents flowing through these can be calculated. By this means it is possible to calculate each of the coupling capacitances exactly.
Consequently, unlike the state of the art, in this method it is unnecessary to exclude the parasitic capacitance between the body and ground either by technical means or instrumentation. The parasitic capacitance can remain, but by making multiple measurements of the measured variables they can be properly taken into account in the calculation of the coupling capacitances of interest, and are therefore not incorporated in the capacitance values of interest. This means that the values of interest can be determined extremely accurately. Furthermore, in a measurement apparatus working according to said method, the technical complexity required to provide dc isolation between the component and ground is no longer necessary.
Apart from the accuracy of the voltage sources and the current measurement, the error susceptibility of the calculated values depends on the coupling capacitances, in particular the ratio of the capacitances to each other. In order to optimise the measurement accuracy it is therefore advantageous that the ac voltages applied to each of the electrodes in the (i+1)-th measurement are dependent on the results of the i-th measurement, where 1xe2x89xa6ixe2x89xa6nxe2x88x921. The measurement voltages for the second measurement of the currents flowing via the electrodes is advantageously set according to this embodiment in such a way, for instance, that the sensitivity of the second measurement is optimum with respect to the capacitance values found from the first measurement.
In order to further increase the accuracy of determining the coupling capacitances it is advantageous to perform more than two measurements, i.e. n greater than 2, and additionally to use statistical methods to analyse the coupling capacitances calculated from the network analysis.